Ask a room of student affairs leaders to define belonging precisely enough to build a program around, and EAB’s research team has found, most will struggle. Even so, the research base is more developed than practitioners often realize, and it points to belonging as one of the more reliable levers colleges have for student retention.
A 2025 critical literature review by researchers at Middlesex University and the University of York, published in the Journal of Learning Development in Higher Education, gathered 118 peer-reviewed studies to map what’s known.
Belonging has a documented connection to 16 different measures of student success, the review found, and retention sits among the strongest and most replicated. Studies using randomized controlled trials and quasi-experimental designs found that increases in a sense of belonging predicted later similar gains in retention, engagement, academic performance, and mental health.
That evidence base is why the Institute for Higher Education Policy (IHEP) has pushed federal policymakers to treat belonging interventions as legitimate completion tools, not soft add-ons. The organization points to concrete numbers. Colleges where large numbers of faculty adopted belonging-focused teaching practices in fall 2020 saw the share of students earning A or B grades rise and the share receiving D’s, F’s, or withdrawing drop substantially.
“Academic setbacks are part of the learning process and do not reflect a student’s potential for success,” IHEP argues in its report on the practice.
What Actually Works
Two experimental studies highlighted by EAB offer some of the clearest evidence that belonging can be built deliberately. Researchers had Black and Hispanic students read and reflect on testimonials from upper-division peers describing how they had worked through their own doubts about belonging. Compared with students who read unrelated material, those who engaged with the testimonials showed a measurably higher sense of belonging and GPA afterward—a low-cost approach that requires just convincing students that self-doubt is common and temporary.
The literature review backs this up from a different angle. Of the 59 factors researchers have tied to student belonging, most have only modest support—a single study here, a qualitative finding there. A handful, though, have shown up repeatedly across institutions and student groups, using rigorous methods: personalized support from staff, group and experiential learning, and programs that proactively address the barriers students expect to face when trying to fit in.
Relationships with academic staff came up more often than any other single factor, but when researchers asked students themselves what makes them feel like they belong, peer relationships edged out staff as the most common answer.
The sheer volume of identified factors—59 of them—suggests belonging is inherently personal rather than one-size-fits-all, according to the review’s authors, David Gilani of Middlesex University and Liz Thomas of the University of York. Both point to the work of scholar Terrell Strayhorn, who has written: “What works for helping one to belong may not work for someone else.”
A Widening Gap
The research is less reassuring when it comes to which students are struggling most. Most report feeling like they belong, per the National Survey of Student Engagement, but belonging isn’t distributed evenly. First-generation, international, LGBTQIA+ students, and those facing financial insecurity or weaker academic preparation, are all less likely to feel welcome and included.
A study published this spring in the International Journal of Environmental Research and Public Health adds a troubling wrinkle. Researchers Joshua Brown of Southern Utah University and Lauren Papp of the University of Wisconsin-Madison tracked 355 students at a large midwestern public university across five surveys, beginning just after they started college and for the next three years.
Contrary to what they expected, a sense of belonging didn’t hold steady over time. “College students’ belonging directly decreased over time,” Brown and Papp write, “and race/ethnicity moderated this change such that being a student from a minoritized racial/ethnic group was associated with steeper drops in belonging.”
Neither Greek life involvement nor a family history of college attendance slowed the decline, which surprised the researchers, who had expected those factors to buffer it; COVID-19’s onset didn’t move the numbers much either.
Brown and Papp offer the theory that as students advance into their majors, their identity may shift from “student at this university” toward “engineer” or “psychologist”—a different problem than the one belonging interventions are usually built to solve, suggesting colleges may need distinct strategies for first-year students versus upperclassmen.
Open Questions
Both research papers are candid about the limits of what’s known. Belonging researchers have used 28 different named measurement scales, plus dozens more built for a single study.
“A risk in student belonging research, given the wide variety of scales used, is that belonging is not being measured consistently,” Gilani and Thomas write, “which hinders comparisons across existing literature.”
Most studies also can’t establish that belonging causes better outcomes rather than simply accompanying them, and 56% of the research came out of the United States, raising questions about whether findings translate elsewhere. The longitudinal study, similarly, comes from a single, majority-White public university. A smaller study at a private university, Brown and Papp note, found belonging held roughly stable over four years rather than declining.
None of that undercuts the basic case for treating belonging as a retention strategy. It does suggest colleges would do well to discern which interventions actually have evidence behind them, rather than assuming any activity branded as belonging-building will work the same way for every student at every stage of their college career.
